A cardinal feature of human cancers is the survival and persistence of residual neoplastic cells - referred to as minimal residual disease - in a presumed quiescent state following the apparently successful treatment of the primary tumor. Ultimately, these cells may re-emerge from their dormant state and resume growth, leading to cancer recurrence. For example, ~30% of breast cancer patients lacking any clinical or histopathological signs of metastasis have disseminated tumor cells present in their bone marrow. These residual neoplastic cells constitute the cellular reservoir from which tumor recurrences invariably arise. Since recurrent breast cancer is typically an incurable disease, the propensity of breast cancer cells to survive in a dormant state and recur is the most important determinant of clinical outcome. Despite the unrivaled clinical importance of these aspects of breast cancer progression, however, little is known about the biology or mechanisms of minimal residual disease, tumor dormancy, and recurrence. Accordingly, understanding the biology of residual tumor cells and elucidating the molecular pathways and cellular processes that contribute to tumor dormancy and recurrence is a critical priority in cancer research. This is the focus of this application. A particular difficulty in studying minimal residual disease, tumor dormancy, and recurrence in breast cancer has been the challenge of identifying and isolating residual neoplastic cells in patients, and the lack of animal models that recapitulate these key features of breast cancer progression. To address this critical gap, we have developed and validated a series of doxycycline-inducible transgenic mouse models for MYC, HER2/neu, Wnt1, and Akt-overexpressing breast cancers that display key features of human breast cancer progression, including minimal residual disease, tumor dormancy, and recurrence. This application will employ a unique set of conditional transgenic mouse models to address fundamental questions regarding the biological properties of residual tumor cells and the pathways that contribute to tumor dormancy and recurrence. The specific aims of this application are to: 1. Determine the contribution of tumor initiating cells to residual neoplastic disease; and 2. Determine the role of par-4, Ssb-1, and the NF-kB pathway in tumor dormancy and recurrence. The first aim will test the hypothesis that residual neoplastic cells that persist in the mammary gland following tumor regression are enriched for cancer stem cells. The second aim will use orthotopic and intact recurrence models to determine the functional contribution of par-4 down- regulation and Ssb-1 up-regulation to mammary tumor recurrence, and the role played by NF-kB pathway activation in this process. Paired patient samples from a neoadjuvant trial will be used to validate the clinical relevance of more studies. Ultimately, understanding the biology of residual neoplastic cells, as well as the mechanisms of tumor dormancy and recurrence, should accelerate the development of more effective approaches to detecting and eradicating dormant tumor cells and preventing breast cancer recurrence.